Mars (Liyao) Gao

Postdoctoral Scholar
Stanford University

Email: marsgao [at] stanford [dot] edu
GitHub | Google Scholar

About me

I am a postdoctoral scholar at Stanford University and a Stanford AI Lab (SAIL) Fellow, working primarily with Professor Emily B. Fox. I received my Ph.D. from the Paul G. Allen School of Computer Science & Engineering at the University of Washington, where I was advised by Professor J. Nathan Kutz.

My research centers on AI for science (AI4S). I develop interpretable and adaptive learning frameworks for scientific discovery, with a particular interest in automated scientific discovery, test-time discovery, and the modeling of complex spatiotemporal systems.

I work at the intersection of AI, deep learning, and the natural sciences, pursuing two complementary goals: using AI to understand and decode complex scientific phenomena, and incorporating scientific knowledge into AI systems to make them safer, more reliable, and more useful in areas such as healthcare, drug discovery, climate science, and clean energy.

Working across diverse scientific domains has also motivated me to study a broader question: how can we design AI models and research workflows that accelerate scientific discovery beyond any single application? My long-term goal is to build reliable AI co-scientists that can work directly with complex scientific data, reason alongside human researchers, and generate meaningful and trustworthy scientific insights.

News

Selected publications

Sparse identification of nonlinear dynamics and Koopman operators with Shallow Recurrent Decoder Networks. Mars L. Gao, Jan P. Williams, J. Nathan Kutz. GitHub Colab Website Youtube Proceedings of the National Academy of Sciences (PNAS).

SENDAI: A Hierarchical Sparse-measurement, EfficieNt Data AssImilation Framework. Xingyue Zhang, Yuxuan Bao, Mars L. Gao, J. Nathan Kutz. Paper GitHub International Conference on Machine Learning (2026).

Bayesian autoencoders for data-driven discovery of coordinates, governing equations and fundamental constants. Mars L. Gao, J. Nathan Kutz. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

Convergence of uncertainty estimates in ensemble and Bayesian sparse model discovery. Mars L. Gao, Urban Fasel, Steven L. Brunton, J. Nathan Kutz. Under review at Transactions on Pattern Analysis and Machine Intelligence (TPAMI).

Deformation Robust Roto-Scale-Translation Equivariant CNNs. Mars L. Gao, Wei Zhu, Guang Lin. Transaction of Machine Learning Research.

On Optimal Early Stopping: Over-informative versus Under-informative Parametrization. Ruoqi Shen, Mars L. Gao, Yi-An Ma. arxiv.